Open access institutions provide opportunities to millions of students who would ordinarily be unable to participate in higher education. Financial aid further increases affordability; however, many students still have substantial unmet need. Innovation and a challenge to traditional thinking about financial aid may be one path to increasing persistence at these institutions. These papers provide early results from rigorous evaluations of three performance-based scholarship programs in MDRC’s Performance-Based Scholarship (PBS) Demonstration.  The programs were designed to help low-income students cover more expenses and provide a financial incentive to make good progress. The scholarships were need-based grants contingent upon meeting performance benchmarks. The first program targeted low-income adult learners in need of remediation at two New York City community colleges. New York City students could earn up to $2,600 over two semesters, and possibly $1,300 over the summer, for maintaining a ‘C’ or better in at least 6 credits per term. The second program targeted low-income parents at three Ohio community colleges. Ohio students could earn up to $900 over two semesters for earning a ‘C’ or better in 6-11 credits per term, and up to $1,800 over two semesters for earning a ‘C’ or better in at least 12 credits per term. The third program targeted low-income freshmen at an open-access university in New Mexico.  New Mexico students could earn up to $1,000 in their first four semesters, for a total of $4,000, by enrolling in a set number of credits (12 in first semester, 15 in subsequent semesters), meeting with advisors, and maintaining a ‘C’ average. The evaluation employs a random assignment research design.  Students who met program eligibility criteria were randomly assigned to two groups: a program group that received the performance-based scholarship, or a control group that received whatever financial aid was available to all students. Random assignment ensured that the motivation levels and personal characteristics of students in all groups were the same at the start of the study.  This presentation will provide program impacts on a wide array of measures including persistence, credits attempted, credits earned, and financial aid.  The early findings suggest that PBS can have a positive effect on students’ course-taking patterns and academic achievement. They also suggest that supplemental funding, not targeted at elite students, could help increase credit accumulation. Lastly, they suggest that PBS can affect whether students enroll full or part-time, ultimately influencing their academic achievement.